Friday, November 04, 2005

Jon Stokes scores C in computing test

Interesting reporting in the NZ Herald on a failed software project at Te Wananga o Aotearoa, based on a leaked report prepared for acting chief financial officer John Mote.

According to Jon Stokes, Te Wananga o Aotearoa spent more than $2.5 million on developing software that was "out-of-date before it was implemented" and has never been used.

It sounds bad, but the story doesn't actually tell us enough to be of any use, apart to add to the general impression that wananga chief executive Rongo Wetere was not up to scratch.

Calling software "out of date" is not particularly informative. Software is not like that container of yoghurt you forgot about in the back of the fridge. Many organisations happily rumble along on software they installed 5, 10, 20 years ago.

Putting on my IT reporter's hat, here is what I could get out of the story.

The system was being developed by a company called InfoQuest, based in Malaysia.

InfoQuest is run by Ganesin Supayah. (I checked the Companies Office record for InfoQuest NZ, and there is someone of that name with a Malaysian address listed as shareholder. The only reference to that name comes up on Google in a list of gongs given out in the state of Agong.)

The software was supposed to be for human resource management and online learning, ie, two completely different systems.

The online computing content was described as "generally poor" and not meeting NZ Qualifications Authority unit standards. I assume the use "computing" in that sentence is redundant.

"Difficulties in co-ordinating development because of language and time zone barriers were also seen as problematic." Hey, outsourcing sucks.

Almost $1 million was spent developing the Human Resource Management System software

The report said the wananga would have to make a "significant commitment" of resources to bring the human resource software up to a usable standard, a cost that must be balanced against spending about $55,000 to buy a complete HR package from a local vendor.

That is the point where I really got concerned about the competence of the people writing this report. Human resources in a polytechnic context, with lots of tutors on fixed time contracts, is complex. The wananga has multiple campuses around the country, which makes it even more difficult. Such systems are not trivial.

The wananga went from a few hundred students in Te Awamutu to being the largest tertiary education provider in the country in a few short years. I would very much doubt it was able to develop adequate IT systems to cope.

Let us not forget that Auckland University, which is far wealthier than the wananga, has spent more $60 million on a student management and HR system. AUT spends millions every year building and maintaining its own systems.

This is not a defence of the wananga. I'm just saying this story actually tells us a lot less than it purports to.

1 Comments:

Blogger Danny Butt said...

Good post - too many people forget Incis etc. There is a history of the sunk software costs of NZ's DIY-ism that needs to be written. From my POV, you buy one of the two or three established packages off-the-shelf or close to it; or you decide to develop in house on open platforms if you want to build your capability to develop software (i.e. if it goes pear-shaped, at least you learn what went wrong). These contracting scenarios are the worst of both worlds.

7:17 AM  

Post a Comment

Links to this post:

Create a Link

<< Home